Online Reviews as Health Data: Examining the Association Between Availability of Health Care Services and Patient Star Ratings Exemplified by the Yelp Academic Dataset
Bibliographic record
Abstract
BACKGROUND: There have been public health interventions that aim to reduce barriers to health care access by extending opening hours of health care facilities. However, the impact of opening hours from the patient's perspective is not well understood. OBJECTIVE: This study aims to investigate the relationship between temporal accessibility of health care services and how patients rate the providers on Yelp, an online review website that is popular in the United States. Using crowdsourced open Internet data, such as Yelp, can help circumvent the traditional survey method. METHODS: From Yelp's limited academic dataset, this study examined the pattern of visits to health care providers and performed a secondary analysis to examine the association between patient rating (measured by Yelp's rating) and temporal accessibility of health care services (measured by opening hours) using ordinal logistic regression models. Other covariates included were whether an appointment was required, the type of health care service, the region of the health care service provider, the number of reviews the health care service provider received in the past, the number of nearby competitors, the mean rating of competitors, and the standard deviation of competitors' ratings. RESULTS: From the 2085 health care service providers identified, opening hours during certain periods, the type of health care service, and the variability of competitors' ratings showed an association with patient rating. Most of the visits to health care service providers took place between normal working hours (9 AM-5 PM) from Sunday to Thursday, and the least on Saturday. A model fitted to the entire sample showed that increasing hours during normal working hours on Monday (OR 0.926, 95% CI 0.880-0.973, P=0.03), Saturday (OR 0.897, 95% CI 0.860-0.935, P<0.001), Sunday (OR 0.904, 95% CI 0.841-0.970, P=0.005), and outside normal working hours on Friday (OR 0.872, 95% CI 0.760-0.998, P=0.048) was associated with receiving lower ratings. But increasing hours during outside normal working hours on Sunday was associated with receiving higher ratings (OR 1.400, 95% CI 1.036-1.924, P=0.03). There were also observed differences in patient ratings among the health care services types, but not geographically or by appointment requirement. CONCLUSIONS: This study shows that public health interventions, especially those involving opening hours, could use crowdsourced open Internet data to enhance the evidence base for decision making and evaluation in the future. This study illustrates one example of how Yelp data could be used to understand patient experiences with health care services, making a case for future research for exploring online reviews as a health dataset.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.044 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.008 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".